Cardinal numerals in rural sign languages: Approaching cross-modal typology ULRIKE ZESHAN, CESAR ERNESTO ESCOBEDO DELGADO, HASAN DIKYUVA, SIBAJI PANDA, and CONNIE DE VOS Abstract This article presents data on cardinal numerals in three sign languages from small-scale communities with hereditary deafness. The unusual features found in these data considerably extend the known range of typological variety across sign languages. Some features, such as non-decimal numeral bases, are unat- tested in sign languages, but familiar from spoken languages, while others, such as subtractive sub-systems, are rare in sign and speech. We conclude that for a complete typological appraisal of a domain, an approach to cross-modal typology, which includes a typologically diverse range of sign languages in addition to spoken languages, is both instructive and feasible. Keywords: cardinal numerals, morphology, numeral bases, numerals, rural sign languages, sign languages 1. Introduction This article discusses primary data on cardinal numerals from three sign lan- guages in small-scale rural communities in Turkey, India, and Mexico. Re- search on such “rural sign languages” is a very recent undertaking, and here we aim to contribute toward what is known about the typology of signed and spoken languages. Sections 1.1 and 1.2 introduce the phenomenon of rural sign languages as compared to the better-documented sign languages of urban deaf communities, as well as presenting an overview of what we know about cardi- nal numerals in urban sign languages, as compared to spoken languages. Sec- tion 2 details the sociolinguistic setting of the target languages, followed by the data analysis in Sections 3 and 4. Importantly, our research results clearly illus- trate the value and feasibility of a cross-modal typology of cardinal numerals, covering a variety of both signed and spoken languages. These implications are explored in Section 5. Linguistic Typology 17 (2013), 357–396 1430–0532/2013/017-0357 DOI 10.1515/lingty-2013-0019 ©Walter de Gruyter Brought to you by | Max-Planck-Gesellschaft - WIB6417 Authenticated | 195.169.108.82 Download Date | 1/6/14 9:27 AM 358 Ulrike Zeshan et al. 1.1. Urban sign languages and rural sign languages Since its inception in the 1960s and 1970s, research on sign languages has focused on communities of deaf people who constitute linguistic and cultural minorities within the larger hearing, non-signing society (e.g., van Cleve 1987, Lucas (ed.) 2001). Since a critical mass of deaf people is needed to give rise to a sign language, these linguistic and cultural minorities are most commonly as- sociated with larger, often urban, populations. They typically include dedicated institutions and infrastructures such as schools for the deaf, deaf associations, etc. They are also characterised by significant communication barriers for deaf individuals, and collective experiences of exclusion and disadvantage vis-à-vis the hearing majority of non-signers, but also a strong sense of “deaf commu- nity” for many deaf individuals (Lane 1992, Padden & Humphries 2006). The sign languages discussed in this article have radically different socio- linguistic and cultural parameters, and research on these languages and their unique setting is very recent (e.g., Zeshan 2010, Meir et al. 2010). They exist in small rural communities with a high incidence of hereditary deafness over several generations, which gives rise to indigenous sign languages separately and independently of urban deaf communities (Groce 1985, Lane et al. 2000). For simplicity’s sake, these languages are collectively called “rural sign lan- guages” here, although this can include cases other than a prototypical single “deaf village” (cf. Section 2). By contrast with urban deaf communities, rural sign language communities are generally characterised by the following fea- tures:1 (i) there is a high incidence of hereditary deafness over several generations; (ii) the sign language is used on a daily basis by the great majority of hearing people in the community; (iii) consequently, there is a much higher degree of integration between deaf and hearing people, and communication barriers are of lesser or no im- portance in the daily lives of deaf people; (iv) there are usually no dedicated institutions for deaf people, e.g., schools or clubs for the deaf. A substantial number of rural sign languages have been identified, mostly lo- cated in developing countries. Over the past decade, sign language linguists have started to document the structures of these sign languages (e.g., Sandler et al. 2005, Nyst 2007, Marsaja 2008, Perniss & Zeshan 2008, Nonaka 2009, de Vos 2012), and it has become clear that rural sign languages can have linguistic structures that differ radically from urban sign languages. The data discussed in this article add to this emerging picture. 1. This is a simplified account, as we also find important differences between rural sign language communities (cf. the descriptions of field sites in Section 2). Brought to you by | Max-Planck-Gesellschaft - WIB6417 Authenticated | 195.169.108.82 Download Date | 1/6/14 9:27 AM Cardinal numerals in rural sign languages 359 All known rural sign languages are endangered, usually because of pres- sure from larger urban sign languages. While the vitality of some urban sign languages is also threatened in the long term, for instance through medical advances resulting in a lower incidence of deafness in industrialised countries, most rural sign languages are in immediate danger of becoming extinct without ever having been properly documented (Zeshan & de Vos (eds.) 2012). 1.2. Numerals in signed and spoken languages Across spoken languages, numerals are among the most widely documented linguistic structures. Both in individual languages and in typological surveys of various sizes, there is a wealth of available data on numeral systems in spoken languages. Typological or comparative studies that discuss numerals include works such as Greenberg (1978), Hurford (1975, 1987), Comrie (1997, 2005a), and Gil (2005). Substantial samples of languages have been used to look at nu- merals across spoken languages, e.g., work reported in Comrie (2005a) with 196 languages, Barriga Puente (1998) with 344 languages, and Hanke (2010) with 281 languages. Across spoken languages, cardinal numerals are particu- larly welldocumented. The same level of documentation is currently not available for sign lan- guages.2 Although the main focus of this article is on little-known sign lan- guages in small rural communities, it is useful to begin with some comments on the structure of numeral systems in various urban sign languages. No sys- tematic typological survey of numeral systems in sign languages has been un- dertaken yet, but information on numerals, in particular cardinal numerals, is available from various sources. Numeral signs feature in many sign language dictionaries, both printed (e.g., Suwanarat et al.1990, Wallin et al. 2006, Tang 2007) and online (e.g., Finnish Association of the Deaf 2003, Johnston et al. 2005, Nederlands Gebarencentrum 2010), and numerals are generally included in sign language teaching materials where such materials exist (e.g., Baker- Shenk & Cokely 1991, Dikyuva & Zeshan 2008). In-depth linguistic studies on aspects of numeral systems in sign languages are much rarer and currently lim- ited to the better-documented sign languages, mostly in industrialised countries (e.g., Fischer 1996, Fuentes & Tolchinsky 2004, Fuentes et al. 2010, Ichida 2005, Mori 2005, Ktejik 2013, Skinner 2007, McKee et al. 2011). Despite this paucity, surveying these resources reveals quite clearly which linguistic structures are common across various urban sign languages around the world, and which are currently unattested in the sign language research 2. A current project in sign language typology at the International Institute for Sign Languages and Deaf Studies, University of Central Lancashire, is aiming to fill this gap. Detailed results are not yet available, but the data from 30 sign languages collected for this on-going project has informed the background discussion here. Brought to you by | Max-Planck-Gesellschaft - WIB6417 Authenticated | 195.169.108.82 Download Date | 1/6/14 9:27 AM 360 Ulrike Zeshan et al. literature. Before we move on to describing the data from rural sign languages, it is useful to summarise common structures found in urban sign languages. These structures are then compared with the findings from primary data in rural sign languages (Sections 3 and 4). A common factor in the classification of numeral systems is the notion of “numeral bases”. For the purpose of this article, numeral bases are conceptu- alised as those numerals on the basis of which other numerals are constructed; that is, in the sense of Comrie (2005b: 207), a base is “that numerical value to which various arithmetical operations are applied”. Following Hanke (2010: 68–69), we distinguish here between additive bases (numeral bases to which other numerals are added) and multiplicative bases (numeral bases which are subject to multiplication).3 Across urban sign languages, as in the great majority of spoken languages, 10 is the most common numeral base. In fact, surveying available materials on urban sign languages has revealed virtually no incidents of sign languages with numeral bases larger than 10.4 The predominance of base-10 numeral systems is paralleled in spoken languages, and there is an evident cognitive link between base-10 numeral systems and the anatomy of the human hands, because “verbal counting has very often, if not always, its origin in physical, or rather manual counting” (Hanke 2010: 72). Nevertheless, bases other than 10 do occur with some regularity across spoken languages, e.g., systems involving the numeral base 20 (vigesimal systems). The sample of 196 spoken languages in Comrie (2005a) includes 10 % of languages with vigesimal systems. For sign languages, such data have so far been unavailable, and the rural sign languages discussed here are the first documented cases of numeral bases higher than 10 in any sign language.